AesSedai/Step-3.7-Flash-GGUF

Quantization (fewer layers) vs stepfun-ai/Step-3.7-Flash. AesSedai/Step-3.7-Flash-GGUF matches part of stepfun-ai/Step-3.7-Flash (747 of its 1471 tensors; the rest of stepfun-ai/Step-3.7-Flash has no counterpart). On the shared part: AesSedai/Step-3.7-Flash-GGUF is a GGUF quantization (Q5_K_M) of the same weights: 235 tensors match exactly and 512 within quantization error.

Tensors changed: Norms 0/193, MLP 0/270, Attention 0/240, other 0/42, lm_head 0/1, Embeddings 0/1

On the Hub · Downloads on Model Pulse